Researchers reduce bias in AI models while preserving or improving accuracy
A new technique identifies and removes the training examples that contribute most to a machine-learning model’s failures.
Learn about artificial intelligence, GPT usage, prompt engineering and other technology news and updates from Land of GPT. The site aggregates articles from official RSS feeds under their original authorship. Each article has a do-follow link to the original source.
A new technique identifies and removes the training examples that contribute most to a machine-learning model’s failures.
Research from the MIT Center for Constructive Communication finds this effect occurs even when reward models are trained on factual data.
Using LLMs to convert machine-learning explanations into readable narratives could help users make better decisions about when to trust a model.
Researchers develop “ContextCite,” an innovative method to track AI’s source attribution and detect potential misinformation.
MIT engineers developed the largest open-source dataset of car designs, including their aerodynamics, that could speed design of eco-friendly cars and electric vehicles.
This new device uses light to perform the key operations of a deep neural network on a chip, opening the door to high-speed processors that can learn in real-time.
Associate Professor Catherine D’Ignazio thinks carefully about how we acquire and display data — and why we lack it for many things.
The MIT Advanced Vehicle Technology Consortium provides data-driven insights into driver behavior, along with trust in AI and advance vehicle technology.
MIT CSAIL researchers used AI-generated images to train a robot dog in parkour, without real-world data. Their LucidSim system demonstrates generative AI's potential for creating robotics training data.
MIT and IBM researchers are creating linkage mechanisms to innovate human-AI kinematic engineering.